Zhou et al. BMC Medicine (2015) 13:154 DOI 10.1186/s12916-015-0391-7

RESEARCH ARTICLE Open Access

hi

CD14 CD16+ monocytes phagocytoseantibody-opsonised Plasmodium falciparuminfected erythrocytes more efficiently than othermonocyte subsets, and require CD16 andcomplement to do soJingling Zhou1, Gaoqian Feng1, James Beeson1,2,6, P. Mark Hogarth1, Stephen J. Rogerson2, Yan Yan3

and Anthony Jaworowski1,4,5*

Abstract

Background: With more than 600,000 deaths from malaria, mainly of children under five years old and caused byinfection with Plasmodium falciparum, comes an urgent need for an effective anti-malaria vaccine. Limited detailson the mechanisms of protective immunity are a barrier to vaccine development. Antibodies play an important rolein immunity to malaria and monocytes are key effectors in antibody-mediated protection by phagocytosingantibody-opsonised infected erythrocytes (IE). Eliciting antibodies that enhance phagocytosis of IE is therefore animportant potential component of an effective vaccine, requiring robust assays to determine the ability of elicitedantibodies to stimulate this in vivo. The mechanisms by which monocytes ingest IE and the nature of the monocyteswhich do so are unknown.

Methods: Purified trophozoite-stage P. falciparum IE were stained with ethidium bromide, opsonised with anti-erythrocyte antibodies and incubated with fresh whole blood. Phagocytosis of IE and TNF production by individualmonocyte subsets was measured by flow cytometry. Ingestion of IE was confirmed by imaging flow cytometry.

Results: CD14hiCD16+ monocytes phagocytosed antibody-opsonised IE and produced TNF more efficiently thanCD14hiCD16- and CD14loCD16+ monocytes. Blocking experiments showed that Fcγ receptor IIIa (CD16) but not Fcγreceptor IIa (CD32a) or Fcγ receptor I (CD64) was necessary for phagocytosis. CD14hiCD16+ monocytes ingestedantibody-opsonised IE when peripheral blood mononuclear cells were reconstituted with autologous serum but notheat-inactivated autologous serum. Antibody-opsonised IE were rapidly opsonised with complement component C3 inserum (t1/2 = 2-3 minutes) and phagocytosis of antibody-opsonised IE was inhibited in a dose-dependent manner by aninhibitor of C3 activation, compstatin. Compared to other monocyte subsets, CD14hiCD16+ monocytes expressed thehighest levels of complement receptor 4 (CD11c) and activated complement receptor 3 (CD11b) subunits.(Continued on next page)

* Correspondence: anthonyj@burnet.edu.au1Centre for Biomedical Research, Burnet Institute, Melbourne, Victoria 3004,Australia4Department of Infectious Diseases, Monash University, Melbourne, Victoria3800, AustraliaFull list of author information is available at the end of the article

© 2015 Zhou et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License(http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium,provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Zhou et al. BMC Medicine (2015) 13:154 Page 2 of 14

(Continued from previous page)

Conclusions: We show a special role for CD14hiCD16+ monocytes in phagocytosing opsonised P. falciparum IE andproduction of TNF. While ingestion was mediated by Fcγ receptor IIIa, this receptor was not sufficient to allowphagocytosis; despite opsonisation with antibody, phagocytosis of IE also required complement opsonisation. Assayswhich measure the ability of vaccines to elicit a protective antibody response to P. falciparum should consider their abilityto promote phagocytosis and fix complement.

Keywords: Malaria, Phagocytosis, Monocyte subsets, Antibodies, Complement, CD16

BackgroundIt is estimated that there are currently more than 200million malaria infections per year, resulting in morethan 600,000 deaths, mainly of children under five yearsold and caused by infection with Plasmodium falcip-arum [1]. In addition, infection with P. falciparum dur-ing pregnancy causes maternal malaria which results inincreased incidence of pre-term births, low infant birthweight and maternal anaemia causing significant mor-bidity and mortality [2, 3].Antibody-mediated effector mechanisms against the

blood stages of the parasite’s life cycle are importantin protection against clinical malaria disease: inmalaria-endemic regions, acquisition of antibodies toblood-stage parasites is associated with protectionagainst death due to severe malaria by five years ofage and with protection against clinical malaria byearly adulthood [4]. Important targets of protectiveantibodies are antigens expressed on the surface ofinfected erythrocytes (IE) [5], and the major target ofthese antibodies is a surface protein known asPfEMP1 [6]. In addition, acquisition of antibodies toantigens exposed on the surface of IE that adhereand accumulate in the placenta, and express thePfEMP1 variant known as Var2CSA, occurs in agravidity-dependent manner and is associated with pro-tection against maternal malaria as well as negative out-comes such as anaemia and low birth weight [7–11].The effector cells most likely to mediate protective

effects of antibodies against circulating blood stageparasites are monocytes, which phagocytose IE [12].They can also accumulate as malaria pigment-ladencells in the placentas of malaria-infected pregnantwomen [13–15]. Monocytes phagocytose IgG-opsonisedIE via Fcγ receptor-mediated mechanisms [16, 17] andsecrete both pro-inflammatory and anti-inflammatorycytokines and growth factors in response to parasiteingestion which may aid in both parasite clearanceand in limiting inflammation [18, 19]. Circulating hu-man monocytes exist as separate subsets which areidentified by their expression of CD14 (the co-receptorfor Toll-like receptor 4 (TLR4) recognition of bacteriallipopolysaccharide) and CD16 (FcγRIIIa: a receptor for

IgG). The current convention is to define three subsetsof human monocytes: classical (CD14hiCD16-), non-classical (CD14loCD16+) and intermediate (CD14hiCD16+)monocytes [20]. The biological properties of these subsetsare governed by differing expression of pattern recogni-tion and chemokine receptors. CD14hiCD16- classicalmonocytes represent the major population in blood,respond strongly to bacterial products via TLR4 andinfiltrate into sites of inflammation in response tothe chemokine CCL2 [21]. CD14loCD16+ non-classicalmonocytes may patrol blood vessel walls and respond toviral ligands via TLR7/8. They express high levels of frac-talkine receptor (CX3CR1) but migrate in response tomultiple chemokines [21]. CD14hiCD16+ intermediatemonocytes may represent a transitional form of the mat-uration of classical monocytes into non-classical mono-cytes and respond strongly to both viral and bacterialligands [22]. The role of different monocyte subsets insettings of parasite infection is not known.While it is recognised that a successful vaccine strat-

egy must generate a robust antibody response to bloodstage parasites, the desired functional activities requiredfor protective immunity are less clear. Evidence is accu-mulating that the ability of antibodies to promoteopsonic phagocytosis of blood stage parasites is an im-portant component of immunity [23–26]. However, themajor cell subsets mediating phagocytosis and theunderlying mechanisms are poorly understood. Thisknowledge may be crucial for the development of highlyprotective vaccines. Monocyte phagocytosis of blood-stage malaria parasites has been previously studied usingperipheral blood mononuclear cells or purified mono-cytes which miss interactions between serum compo-nents, uninfected erythrocytes and the phagocyte, andhave not usually considered individual monocyte subsetresponses. Here we use a whole blood phagocytosis assay[27] to show for the first time that CD14hiCD16+ inter-mediate monocytes have a much greater phagocytic ac-tivity towards trophozoite stage malaria parasites thanother monocyte subsets. Our results uncover an essen-tial role for Fcγ receptor IIIa (CD16a) and show thatcomplement opsonisation is required for IgG-mediatedphagocytic uptake under physiological conditions and

Zhou et al. BMC Medicine (2015) 13:154 Page 3 of 14

contributes to the high activity of intermediate mono-cytes against IE.

MethodsEthics approvalBlood was obtained by venepuncture with informedconsent from healthy volunteers without a history ofmalaria infection using protocols approved by The AlfredHospital Research and Ethics Unit. Five serum sampleswith high IgG reactivity to CS2 IE were pooled from sam-ples collected from pregnant women in an endemic regionof Papua New Guinea [28]. All women gave informedwritten consent and ethics approval was provided by theMedical Research Advisory Committee, PNG).

Parasite culture and purificationP. falciparum laboratory lines CS2 [29] and E8B [30]were grown in human erythrocytes (group O, Rh+,Australian Red Cross Blood Service) at 37 °C with 5 %CO2 suspended in RPMI-HEPES medium supple-mented with 50 μg/ml hypoxanthine, 25 nM NaHCO3,20 μg/ml gentamicin, 5 % heat inactivated pooled hu-man serum and 5 % Albumax. Gelatin enrichment ofknob-expressing IE was performed weekly, and IEwere synchronised weekly by resuspension of culturepellets in 5 % sorbitol in water to lyse trophozoite andschizont IE. Mature pigmented trophozoite stage IEwere enriched by centrifugation over Percoll gradientsto >80 % purity as assessed by counting Giemsa-stainedthin blood smears by microscopy.

Infected erythrocyte labeling and opsonisationIE were opsonised at a concentration of 5 x 107 tropho-zoites/mL for 30 minutes at room temperature withrabbit anti-human erythrocyte antibodies (Cappel, MPBiomedicals, LLC; Santa Anna, CA, USA) using a sub-agglutinating 1/800 dilution of antibody in phosphate-buffered saline (PBS). In some experiments CS2 IE wereopsonised with 20 % human immune serum from a poolof sera prepared from pregnant woman with placentalmalaria enrolled in the VT cohort in Papua New Guinea[31]. Opsonised cells were washed in fluorescence-activated cell sorting (FACS) wash buffer (PBS, 2 % newborn calf serum) and resuspended in PBS (2 x 108/mL)then stained with 10 μg/mL ethidium bromide (EtBr) for30 minutes at room temperature. Following labelingwith EtBr, cells were washed three times with cold FACSwash buffer and used immediately.

Whole blood phagocytosis assayA sample of 5 mL whole blood was collected fromhealthy volunteers into lithium heparin blood collectiontubes by venepuncture and analysed within two hours ofcollection. Aliquots of 50 μL whole blood were placed in

polypropylene FACS tubes, then 1 x 107 EtBr-labeled IEwere added. This is a ratio of approximately 200 IE perperipheral blood mononuclear cell. Cells were incubatedfor 30 minutes at 37 °C, or on ice as a control. Afterphagocytosis, cells were lysed with 3 mL 0.2 % ammo-nium chloride for five minutes at 22 °C, then washedwith 3 mL cold FACS wash buffer. Supernatant was re-moved and cells were resuspended in 100 μL PBS. Cellswere stained with antibodies for 30 minutes on ice,washed, then fixed with 2 % formaldehyde and analysedimmediately by flow cytometry using a FACS Canto IIflow cytometer (BD Biosciences, San Jose, CA, USA).Monocyte subsets were identified by staining with antiCD14 APC (M5E2, BD Biosciences) and CD16 FITC(3G8, BD Biosciences), and phagocytosis was determinedby measuring EtBr fluorescence in the PE channel. Gateswere set using samples incubated with unopsonised,EtBr-stained IE at 37 °C. Flow data were analysed usingFlowJo (version 8, Tree Star Inc.). For blocking experi-ments, blood was pre-incubated with the indicated con-centrations of blocking antibodies for 30 minutes at 4 °Cbefore the addition of IgG-opsonised IE. The antibodiesused were 3G8 (in house (MH): blocking antibody forCD16), Fab fragment of IV.3 (in house (MH): blockingAb for CD32a), 10.1 (Santa Cruz Biotechnology, Dallas,TX, USA : blocking Ab for CD64), H1-111 (Biolegend,San Diego, CA, USA: blocking antibody for CD11a),Bear-1 (Abcam, Cambridge, UK: blocking antibody forCD11b) and clone 3.9 (Biolegend: blocking antibody forCD11c). The antibody used to measure C3b depositionof RBC was FITC-conjugated goat fraction to humancomplement C3, (MP Biomedical 0855167). To measurethe effect of inhibiting complement activation on phago-cytosis, compstatin (R&D Systems, Minneapolis, MN,USA) was added to whole blood, from stock solutions(2 mg/ml) prepared in PBS, to a final concentration of0–50 μM, and incubated for 15 minutes on ice beforeaddition of iRBC and transfer to 37 °C to measurephagocytosis.

Monocyte phenotypingAliquots of peripheral blood mononuclear cells (PBMC)from malaria-naive healthy individuals recruited inMelbourne were incubated with predetermined saturat-ing concentrations of the relevant antibodies. Monocyteswere gated using forward and side scatter and mono-cyte subsets identified with CD16 PE Cy7 (3G8, BDBiosciences) and CD14 BV510 (M5E2, Biolegend) orAPC (M5E2, BD Biosciences). antibodies used were:CD16 PE Cy7 (3G8, BD Biosciences), CD32a (IV.3 bi-otinylated Fab fragment + streptavidin APC), CD32b (inhouse (MH), 63X-21 biotinylated whole IgG + streptavidinAPC) [32], CD64 PerCP 5.5 (10.1, Biolegend), CD11aAlexa-488 (HI111, Biolegend), CD11b APC (ICRF44,

Zhou et al. BMC Medicine (2015) 13:154 Page 4 of 14

Biolegend), activated CD11b (CBRM1/5-FITC, Biolegend),CD11c V450 (B-ly6, BD Biosciences), CD35 FITC (E11,Biolegend).

Intracellular TNF measurementsPhagocytosis of infected red blood cells (iRBC) was per-formed using 100 μL aliquots of whole blood as de-scribed above except that iRBC were not labeled withEtBr. A total of 20 μg/mL brefeldin A and 10 μM mon-ensin was added and the cells incubated for four hoursat 37 °C, stained with CD14 APC and CD16 PE Cy7(30 minutes on ice), then permeabilised (Perm/WashBuffer 1, BD Biosciences). After 10 minutes on ice, cellswere stained with αTNF phycoerythrin (PE) (Mab11, BDBiosciences) for 30 minutes, washed and fixed.

Imaging flow cytometryIE were stained with PKH26 (Sigma-Aldrich, Castle Hill,NSW, Australia) according to the manufacturer’s instruc-tions. Aliquots of whole blood (50 μL) were incubatedwith 5 x 106 PKH26-stained CS2-IE for 15 minutes, andprocessed as above for whole blood phagocytosis exceptthat cells were stained with CD14 Pacific Blue (M5E2,Biolegend) and CD16 PE Cy5 (3G8, BD Biosciences).Samples were acquired using an ImageStream 100imaging flow cytometer and analysed using IDEASsoftware.

ResultsCD14hiCD16+ monocytes in whole blood phagocytose IEmore efficiently than other monocyte subsetsWhole blood obtained from an individual with no previ-ous history of malaria infection was incubated withEtBr-stained CS2-IE, opsonised or not opsonised withrabbit anti human RBC IgG, and phagocytosis analysedby flow cytometry. The CS2 parasite isolate was chosenboth for its relevance to pregnancy-associated malariaand because the lack of CD36 binding potentiallyminimises the level of non-opsonic phagocytosis.Events within the broad monocyte gate defined by for-ward and side scatter were analysed on a CD14 versusCD16 dot plot to define the three monocyte subsetsCD14hiCD16- “classical”, CD14hiCD16+ “intermediate”and CD14loCD16+ “non-classical” monocytes (Fig. 1a).The extent of IE phagocytosis by each monocyte subsetwas determined from the intensity of EtBr fluorescenceand compared to the 4 °C negative control. There waslittle or no phagocytosis of unopsonised IE by anymonocyte subset (Fig. 1a, right, upper panels). Opson-isation with IgG increased phagocytosis of IE, particu-larly by intermediate monocytes (Fig. 1a, right lowerpanels). Surprisingly, we detected little IE ingestionin the CD14hiCD16- or the CD14loCD16+ subsets.CD14hiCD16+ monocytes showed much higher

phagocytosis of both IgG-opsonised and non-opsonised IE with the CD14loCD16+ monocyte sub-set showing the least amount of activity (Fig. 1b).These differences were not due to generally higherphagocytic activity of CD14hiCD16+ compared to othermonocyte subsets since when unopsonised Escherichiacoli was used as a target the classical subset showed thehighest degree of phagocytosis (Additional file 1: FigureS1). Nor was it due to a greater ability of CD14hiCD16+monocytes to phagocytose particles of the size of erythro-cytes (7 μm diameter) since when IE were incubated withisolated PBMC instead of whole blood the classicalmonocyte subset ingested IgG-opsonised IE to a similarextent (Fig. 1c). Since we used rabbit anti-humanerythrocyte IgG to opsonise IE to high levels, we nextconfirmed that CD14hiCD16+ monocytes showed en-hanced phagocytosis of IE opsonised with human IgG. IEwere opsonised with a pool of immune sera from womenwith placental malaria who had a high titre of antibodiesrecognising the CS2 isolate; CD14hiCD16+ monocyteswere again the only subset to substantially phagocyt-ose parasites (Fig. 1d). CS2 is a P. falciparum linethat expresses Var2CSA. To determine if the in-creased phagocytic capacity of CD14hiCD16+ mono-cytes was specific to this parasite strain, weincubated whole blood with E8B-IE. E8B is a malariastrain that expresses a mixture of var genes that, in con-trast to CS2, promote binding to CD36 and ICAM-1 [33,34]. CD14hiCD16+ monocytes were the only monocytesubset to efficiently ingest IgG-opsonised E8B-IE(Fig. 1e) indicating that the specificity forCD14hiCD16+ monocytes is independent of thePfEMP-1 type. More monocytes ingested IE whenPBMC preparations were used in the phagocytosisassay compared to whole blood (Fig. 1c c.f 1d). This wastrue for both CD14hiCD16+ monocytes (median phago-cytosis = 34.4 c.f. 10.4, p = 0.02) and forCD14hiCD16- monocytes (median phagocytosis = 47.9 c.f.4.22, p = 0.0007).

The higher phagocytic ability of CD14hiCD16+ monocytes,that is evident in whole blood but not PBMC, is not dueto lower inhibition by RBC or plasmaSerum and uninfected RBC have been reported toinhibit phagocytosis of iRBC [35]. To test whether unin-fected RBC inhibit phagocytosis of IE, and whether thisinhibition is less for CD14hiCD16+ monocytes, we incu-bated PBMC with titrated amounts of group O-negativeRBC then performed phagocytosis. RBC inhibitedphagocytosis by monocytes from all three subsetspresent in PBMC when added at 25-200x the number ofPBMC (Additional file 2: Figure S2A). The maximumratio of RBC to PBMC used in this experiment was200:1. This equates to a concentration of 1 x 109/ml

Fig. 1 (See legend on next page.)

Zhou et al. BMC Medicine (2015) 13:154 Page 5 of 14

(See figure on previous page.)Fig. 1 CD14hiCD16+ intermediate monocytes phagocytose IE more efficiently than other monocytes. a Whole blood was incubated with EtBr-labelledCS2-IE for 30 minutes then uningested RBC removed by hypotonic lysis and washing. Cells were stained with anti-CD14 and CD16, monocytes gatedusing forward and side scatter then subsets defined as classical (C: CD14hiCD16-), intermediate (IM: CD14hiCD16+) and non-classical (NC: CD14loCD16+)as shown. Histograms show EtBr staining of the three subsets incubated at 37 °C (red histograms) or 4 °C (blue histograms) with unopsonised (IE, top)or opsonised (IgG-IE, bottom) IE. b Phagocytosis using blood from eight separate donors. Whole blood was incubated as in a with unopsonised CS2-IE(left hand panels; IE) or CS2-IE opsonised with rabbit anti-human RBC antibody (right hand panels; IgG-IE) as indicated. c Phagocytosis bymonocyte subsets of IE opsonised with rabbit anti human RBC was measured using PBMC prepared from four separate donors (left handpanels). Phagocytosis of IE opsonised with pooled human immune serum was measured using PBMC prepared from six separate donors(right hand panels). d Phagocytosis of unopsonised CS2-IE (left hand panels; IE) and CS2-IE opsonised with pooled human immune serum(right hand panels; IgG-IE) was measured in a whole blood assay as in a using blood from nine separate donors. e Phagocytosis using bloodfrom six separate donors. Whole blood was incubated as in a with unopsonised E8B-IE (left hand panels; IE) or E8B-IE opsonised with rabbitanti-human RBC antibody (right hand panels; IgG-IE) as indicated. Background phagocytosis measured at 4 °C was subtracted from all datapoints. The percent phagocytosis by intermediate (IM) monocytes was compared using pairwise comparisons in each case (b-e) with eitherthat by classical (C) monocytes or non-classical (NC) monocytes, as indicated. Differences between groups were assessed using Wilcoxonmatched pairs signed rank test: * p < .05, ** p <0.01. EtBr ethidium bromide, IE infected erythrocytes, PBMC peripheral blood monocytes;RBC red blood cells

Zhou et al. BMC Medicine (2015) 13:154 Page 6 of 14

which is lower than that found in normal humanblood (4-6 x 109/ml). To test whether componentspresent in human plasma inhibit monocyte phagocyt-osis, PBMC were incubated with varying concentra-tions of autologous heat-inactivated plasma thenphagocytosis measured as above. Plasma inhibitedphagocytosis of IgG-opsonised IE by all three mono-cyte subsets and, in particular, of CD14hiCD16+ andCD14hiCD16- monocytes equally (Additional file 2:Figure S2B). Thus, uninfected RBC and humanplasma both inhibit phagocytosis of iRBC but theirpresence does not account for the higher phagocyt-osis by CD14hiCD16+ monocytes observed in wholeblood.

Ingestion of IE by CD14hiCD16+ monocytes wasconfirmed using imaging flow cytometryWe next incubated whole blood with opsonised IE for15 minutes, lysed the uningested erythrocytes, thenanalysed single cells in focus using imaging flow cy-tometry (Additional file 3: Figure S3). A shorter timewas used to allow ingestion without substantial diges-tion of the IE. Bright field images of intermediatemonocytes confirmed the presence of ingested para-sites and the co-localisation of CD16 around thephagosome. Manual counting of ingested parasitesusing approximately 300 randomly selected bright fieldimages within their respective gates confirmed thehigher phagocytic index (PI) of CD14hiCD16+ mono-cytes (33 parasites ingested per 324 monocytes ana-lysed or PI = 10.2 parasites ingested per 100monocytes) compared to CD14hiCD16- monocytes(13/288 or PI = 4.51). The lower extent of phagocytosis

in this experiment compared to the experimentsdepicted in Fig. 1b is due to a lower concentration ofIE and the shorter time used.

Complement opsonisation is absolutely required forphagocytosis of IgG-opsonised IE in whole bloodSince the higher phagocytic activity by intermediatemonocytes was observed only when IE were added towhole blood, but not to PBMC preparations, we rea-soned that opsonins apart from IgG, such as comple-ment components, might contribute to this activity.We, therefore, centrifuged heparinised whole blood,washed and reconstituted the cells to the originalblood volume using either autologous serum or heat-inactivated autologous serum (collected in a separateserum tube at the same time as blood collection), thenmeasured phagocytosis of CS2 IE. Phagocytosis wasnot affected by washing and reconstituting the bloodcells in normal serum, but was abolished when heat-inactivated serum was used (Fig. 2a). These data sug-gest that complement opsonisation occurs during the30 minute incubation of IE with whole blood and thatthis opsonisation is essential for efficient phagocytosisof IgG-opsonised IE by CD14hiCD16+ monocytes. Toverify this, we added purified, IgG-opsonised CS2-IE toheparinised plasma for various times at 37 °C and mea-sured deposition of C3b onto the opsonised IE by flowcytometry. There was minimal C3b bound to IE at 0time (Fig. 2b, left and middle panels; dark grey histo-grams), or after 30 minutes in the absence of IgG op-sonisation (Fig. 2b, left panel: light grey histogram),but considerable deposition onto IgG-opsonised IEafter 30 minutes (Fig. 2b, middle panel: light greyhistogram). C3b was deposited onto IE with a half time

Fig. 2 Complement opsonisation is required for monocyte phagocytosis of antibody-opsonised IE a Monocyte phagocytosis of CS2-IE (Unop) orCS2-IE opsonised with rabbit anti-human RBC antibody (OP) was determined using whole blood, whole blood reconstituted to its originalvolume with autologous heat-inactivated plasma (HI plasma) or with autologous plasma (plasma) as indicated. Phagocytosis by classicalCD14hiCD16- (left hand panel), intermediate CD14hiCD16+ (middle panel) and non-classical CD14loCD16+ (right hand panel) monocytes weremeasured. Data represent mean (sem) of independent experiments using blood from three separate donors. Differences between conditions wereassessed by one-way ANOVA using Tukey’s test for multiple comparisons. b Unopsonised CS2-IE or CS2-IE opsonised with rabbit anti-human RBCantibody were added to heparinised plasma for 0 and 30 minutes at 37 °C, stained with antiC3, and RBC analysed by flow cytometry. His-tograms represent C3 staining at 0 time (dark grey histogram) or 30 minutes (light grey histogram). Right hand panel represents C3 stain-ing of CS2-IE opsonised with rabbit anti-human RBC antibody after incubation in heparinised plasma for the indicated times at 4 °C (solidblack circles) or 37 °C (open circles). c Compstatin (R&D Systems) was added to whole blood from stock solutions dissolved in PBS at theindicated final concentrations then phagocytosis of CS2-IE opsonised with rabbit anti-human RBC antibody by intermediate (open circles)or classical monocytes (solid black circles) determined (left hand panel) or non-classical monocytes (open squares) determined (right handpanel). The absolute values of phagocytosis by non-classical monocytes were very low and hence these data are plotted separately. Data representmean (sem) of independent experiments using blood from three separate donors. ANOVA analysis of variance, IE infected erythrocytes, RBC redblood cells, sem standard error of the mean

Zhou et al. BMC Medicine (2015) 13:154 Page 7 of 14

of 2.7 minutes (Fig. 2b, right panel). The fact that com-plement deposition required IgG opsonisation suggeststhat complement is fixed under these conditions

primarily by the classical pathway. To determine thesignificance of complement to IE phagocytosis bymonocytes in whole blood, we next examined the

Fig. 3 CD14hiCD16+ monocytes produce more TNF compared to other monocytes in response to IE. a Representative histograms showingintracellular TNF staining of monocytes four hours after addition of CS2-IE (IE, left hand panels) or CS2-IE opsonised with rabbit anti-human RBCantibody (IgG-IE, right hand panels). Grey histograms: 4 °C controls, red histograms: 37 °C. b Median (IQR) of intracellular TNF expression in classical(C; solid black circles), intermediate (IM; open circles) and non-classical (NC; open squares) monocytes from four independent experiments usingblood from separate donors. IE infected erythrocytes, IQR interquartile range

Zhou et al. BMC Medicine (2015) 13:154 Page 8 of 14

Zhou et al. BMC Medicine (2015) 13:154 Page 9 of 14

effect of an inhibitor of C3 activation, compstatin.Compstatin inhibited phagocytosis of IgG-opsonisedIE by both CD14hiCD16- and CD14hiCD16+ mono-cytes (Fig. 2c) showing that even when opsonised withIgG, complement opsonisation is required for efficientphagocytosis of IE. In these experiments phagocytosisby CD14loCD16+ monocytes was very low and wasthus excluded from this analysis.

CD14hiCD16+ monocytes produced TNF in response toIgG-opsonised IECD14hiCD16+ monocytes respond to phagocytosis ofbacterial pathogens by producing pro-inflammatory cy-tokines such as TNF [22]. Since this is thought to be re-quired for both effective immunity to malaria and forimmunopathogenesis we determined whether intermedi-ate monocytes produce TNF in response to IE. Wholeblood was incubated with unopsonised and opsonisedCS2-IE for four hours, then intracellular TNF measuredby flow cytometry for all three subsets. There was noTNF production in response to unopsonised IE (Fig. 3a,left hand panels). Both CD14hiCD16- and CD14hiCD16+monocytes produced TNF following addition of opso-nised parasites, with more of the CD14hiCD16+ subsetproducing TNF in agreement with their greater phago-cytic potential, although this difference did not reach sig-nificance, whereas CD14loCD16+ monocytes producedvery little (Fig. 3a, right hand panels and Fig. 3b).

Fcγ and complement receptors required for phagocytosisof IE in whole bloodWe next phenotyped monocytes from nine independentdonors to determine how the subsets differ with respectto expression of receptors involved in binding andphagocytosis of complement and IgG-opsonised targets.Of the phagocytic Fcγ receptors, CD14hiCD16+ mono-cytes expressed significantly higher levels of Fcγ receptorIIa, CD32a, compared to the other subsets (Fig. 4a). Thelevel of the inhibitory Fcγ receptor, CD32b, was alsohighest in this subset, although this receptor appeared tobe expressed at much lower levels than CD32a. Withrespect to the phagocytic complement receptors, theCD14hiCD16+ monocytes expressed the highest levelsof the α chain of CR4, CD11c. Of interest, however,was the observation that although CD14hiCD16+monocytes expressed levels of CD11b (the α chain ofCR3) that were intermediate between expression onthe CD14hiCD16- and CD14loCD16+ subsets, theyexpressed the highest levels of activated CD11b suggest-ing that inside-out signaling required for CR3 activationwas relatively stronger in this subset. CD14hiCD16+monocytes also expressed the highest levels of the αchain (CD11a) of the adhesion molecule LFA-1. SinceCD32a is the only Fcγ receptor expressed most highly on

CD14hiCD16+ monocytes relative to other subsets wereasoned that it may have a unique role in phagocytosisof IgG-opsonised IE. We, therefore, used blockingantibodies to determine which Fcγ receptors are requiredfor phagocytosis. Aliquots of whole blood were pre-incubated for 30 minutes with blocking antibodies spe-cific for CD16, CD32a and CD64, then IgG-opsonisedCS2 IE were added and phagocytosis measured after30 minutes. The blocking antibody specific for CD16,3G8, inhibited phagocytosis by CD14hiCD16+ andCD14loCD16+ monocytes by approximately 90 % at10–20 μg/mL (Fig. 4b upper panel) but, as expected,had no effect on phagocytosis by CD14hiCD16-monocytes which do not express CD16. This inhibitionwas confirmed using whole blood from three individualdonors incubated with 10 μg/mL blocking antibodies(Fig. 4b lower panel). In contrast, blocking antibodiesspecific for CD32a, IV.3, and for CD64, 10.1, had no ef-fect on phagocytosis by any subset despite these recep-tors being expressed on all three subsets. Thus, CD16,but not CD32a or CD64 is necessary for phagocytosis ofIgG-opsonised IE in whole blood. However, the fact thatnon-classical monocytes express CD16 but phagocytoseIE poorly shows that CD16 expression is not sufficient.Since complement opsonisation was also necessary forphagocytosis of IgG-opsonised IE in whole blood, we in-vestigated the effect of blocking antibodies to the phago-cytic complement receptors CR1, CR3 and CR4 as wellas antibodies to LFA-1. Antibodies to the α chain of CR3(CD11b) and CR4 (CD11c) showed minimal inhibition at10 μg/mL but blocked more efficiently at higher concen-trations (Fig. 4c). Anti CD11a did not inhibit phagocyt-osis by any monocyte subset.

DiscussionUsing a whole blood phagocytosis assay we studied theproperties of phagocytes under conditions resemblingthose in vivo as closely as possible. We show that al-though the CD14hiCD16- classical and CD14hiCD16+intermediate monocyte subsets efficiently phagocytoseIE in PBMC preparations, only the CD14hiCD16+ subsetdoes so in whole blood. While the CD14hiCD16+ subsetis more efficient at phagocytosis on a per cell basis, thegreater number of CD14hiCD16- monocytes suggeststhat they may also phagocytose significant numbers ofIE in vivo. We show for the first time that phagocytosisof IgG-opsonised IE required complement opsonisationand was strongly inhibited by inhibitors of complementactivation. Antibody blocking experiments showed thatin whole blood phagocytosis required expression of theFcγ receptor CD16, but not CD32a or CD64. Thus,CD16 is necessary but not sufficient for phagocytosissince non-classical monocytes, which also express CD16,failed to phagocytose IE efficiently. This is likely due to

Fig. 4 (See legend on next page.)

Zhou et al. BMC Medicine (2015) 13:154 Page 10 of 14

(See figure on previous page.)Fig. 4 Expression of Fcγ and complement receptors on monocyte subsets and the effect of blocking antibodies on phagocytosis by individualsubsets. a Expression of Fcγ and complement receptors on monocytes was determined by whole blood staining. Black bars: C, white bars: IM,grey bars: NC. Bars represent mean (sem) of MFI using blood from nine separate donors (eight for CD32b). aCD11b refers to activated CD11bdefined by the epitope recognised by the CBRM 1/5 monoclonal antibody. Differences between subsets were assessed using Wilcoxon’s matchedpairs signed rank test. * p <0.05, ** p <0.01. b Whole blood was incubated for 30 minutes at 4 °C with the indicated concentrations of eachblocking antibody before addition of CS2-IE and determination of phagocytosis. Representative dose response curves from four independentexperiments, of inhibition of phagocytosis by closed black circles), IM monocytes (open circles) and NC monocytes (open squares) are shownin the upper panels and data (mean, sem) from experiments with whole blood from three separate donors conducted using 10 μg/mL of eachblocking antibody are shown in the lower panels. c Effect of blocking LFA-1 (CD11a), CR3 (CD11b) and CR4 (CD11c) on phagocytosis of IgGopsonised CS2-IE. Upper panels show dose response of the indicated blocking antibodies and lower panels show aggregate data (mean, semfrom n =3 independent experiments). Symbols are the same as in a and b. C classical, IM intermediate, N non-classical

Zhou et al. BMC Medicine (2015) 13:154 Page 11 of 14

their lower expression of CR1, 3 and 4 and lower levelsof activated CR3. Complement opsonisation of IE oc-curred rapidly in situ during the assay, primarily via theclassical pathway as we did not detect complement de-position on IE in the absence of IgG opsonisation. Inwhole blood of children with malaria infection, IE haveC3b and C4 deposited on their surface [36] but it is notclear whether the limiting factor for efficient phagocyt-osis is the extent of complement fixation, opsonisationwith IgG or both. Our findings show that evaluation ofimmunity in vaccine trials must also take into accountthe ability of the elicited antibodies to fix complement aswell as to promote phagocytosis. The use of the CS2strain in this study gives relevance to anti Var2CSA vac-cines currently under development for pregnancy-associated malaria and currently funded through toclinical trials. It is essential to understand how anti-bodies to Var2CSA function in order to design a vaccinewith maximum efficacy and to evaluate responses insuch trials.We propose that co-operation between CD16 and

complement receptors are required in whole blood forphagocytosis of IgG opsonised IE. Multiple complementreceptors are probably involved although a likely partnerfor CD16 is CR3 (CD11b/CD18) which was more acti-vated on intermediate monocytes than on other mono-cyte subsets, and partial inhibition of phagocytosisoccurred with the blocking antibody Bear-1. CD16 isknown to interact with CR3 on monocytes, enhancingits ability to bind iC3b [37]. The selective role of CD16expressed on monocytes may be due to specific inter-action with CR3 and/or signaling differences between itand CD32a. This latter could be caused either by differ-ences in the ITAM motifs located on the FcRγ signalingprotein associated with CD16a and that in the cytoplas-mic domain of CD32a, or by signaling pathways acti-vated following phosphorylation of the cytoplasmicdomain of CD16 [38].Our data, primarily obtained using IE opsonised with

rabbit anti-human RBC antibody, were verified using IEopsonised with human immune serum. Although we

did not explore the human IgG isotypes required,studies have shown that anti-malaria antibodies pro-moting phagocytosis are mainly cytophilic IgG1 andIgG3 [39, 40]. We used two separate laboratory-derived P. falciparum lines, CS2 and E8B which wereboth more efficiently ingested by CD14hiCD16+monocytes. These parasite lines express PfEMP1 ad-hesion molecules on the surface of the IE that bindwith different ligand specificities. Thus, CS2 bindspreferentially to chondroitin sulfate A [29] while E8Bbinds to both CD36 and ICAM-1 [34]. Our observa-tions, therefore, rule out involvement of these receptorsin the increased ability of CD14hiCD16+ monocytes inphagocytosis.Imaging flow cytometry verified that under the condi-

tions of our experiments ingestion of IE and not surfacebinding were measured. This is important since weobserved that fragments of bound RBC may remain at-tached to monocytes following hypotonic lysis of theuningested RBC which can lead to unacceptably highbackgrounds when membrane stains are used to labeltarget cells. We found that labelling of the parasite DNAwithin IE using EtBr was the best approach to use formeasurement of IE phagocytosis, but it has the disad-vantage that analysis must be performed within 30–60minutes to avoid loss of the EtBr stain. This lessens theutility of the whole blood assay used here in clinicalsettings.Our data have uncovered an important role of CD16

expressed on monocytes in responses to IE. Monocytesexpress the transmembrane receptor CD16a in contrastto neutrophils, which express the GPI-linked receptorCD16b. Although CD16a and CD16b have nearly identi-cal extracellular domains, they are encoded by separategenes [41]. Polymorphisms of CD32a and CD16b are as-sociated with malaria severity [42–45] that may reflectthe ability of splenic macrophages and neutrophils, re-spectively, to clear opsonised parasites or, in the case ofassociations with severe anemia, to ingest uninfectedRBC. Several polymorphisms in CD16a affect either af-finity for cytophilic IgG subclasses [46] or monocyte

Zhou et al. BMC Medicine (2015) 13:154 Page 12 of 14

expression [47]; however, to our knowledge no studieshave attempted to associate these or other CD16a poly-morphisms with malaria severity or vaccine responses.Given the absolute requirement of IE phagocytosis forCD16 expressed on monocytes revealed herein, suchstudies are indicated.The ability of monocytes to phagocytose IE in whole

blood is decreased relative to that of PBMC prepara-tions, highlighting the caution that must be employedwhen interpreting results using PBMC or purifiedmonocytes in phagocytosis assays. This may reflect thepresence of inhibitory factors in serum or the large num-ber of uninfected RBC in whole blood. Uninfected redblood cells inhibited ingestion of IgG-opsonised tropho-zoite stage IE consistent with observations of othersusing unopsonised schizont-stage IE [35]. We foundsimilar inhibition of all three monocyte subsets, however,suggesting that this is unlikely to underlie their differingphagocytic ability. Interestingly, human RBC membranescontain an inhibitory factor (phagocytosis-inhibitory fac-tor, PIF) which appears to affect CR3 conformation andinhibits ingestion of both C3bi and IgG opsonised latexbeads [48]. We also provide evidence that phagocytosisby classical and intermediate monocytes in whole bloodis lower than that by these cells in PBMC because of aninhibitory effect of soluble plasma components. It is pos-sible that binding of IgG present in serum to Fcγ recep-tors may contribute to this, since removal of IgG usingprotein G sepharose beads reduced inhibition by addedplasma (Additional file 4: Figure S4).Elevated numbers of CD14hi monocytes which co-

express the chemokine receptors CCR2 and CX3CR1have been associated with lower parasitemia and in-creased ADCI activity in P. falciparum-infected individ-uals with uncomplicated malaria [49]. These cells maydefine a specific intermediate monocyte population withan important protective role against blood stages of theparasite. It would be of interest to directly compareADCI activity of different monocyte subsets and thephenotype of monocytes with high phagocytic and highADCI activity to determine if the same populations areinvolved. A selective role for CD14hiCD16+ monocytesin the response to P. vivax-infected reticulocytes has re-cently been published [49]. This report differs from ourstudy in that increased phagocytosis by CD14hiCD16+monocytes was observed using PBMC preparations incu-bated for long times (four hours) with purified, infectedreticulocytes. In our hands, discrete monocyte subsetscannot be identified after four hours incubation of wholeblood with phagocytic targets since CD16 expression onCD16+ monocytes is decreased. A second difference be-tween the two studies is that Antonelli et al. [50] usedPBMC prepared from patients with active P.vivax infec-tion which were shown to be in an activated state,

whereas we have studied responses of individuals withno history of malaria infection and who therefore repre-sent individuals at risk of primary infection. The differ-ences between the two studies may also be due tothe different phagocytic targets (P. falciparum-infectederythrocytes versus P. vivax-infected reticulocytes) usedin the two studies. Nevertheless, both studies point to animportant role for CD14hiCD16+ monocytes in the con-trol and response to blood-stage malaria infection.In summary, our data show a special role for

CD14hiCD16+ monocytes in the phagocytosis oftrophozoite-stage P. falciparum IE. The use of wholeblood assays to measure phagocytosis in place of purifiedPBMC or monocytes reveals that efficient phagocytosisrequires both IgG opsonisation and complement compo-nents in situ. Evaluation of vaccine candidates is increas-ingly employing functional assays such as phagocytosisassays to determine correlates of immunity. Ideally suchassays should use fresh whole blood, although this maynot be feasible in some field settings. Our data suggestthat care must be taken in interpreting the results of as-says using purified cells and that the ability of antibodiesto fix complement are likely to be as relevant as opsonicactivity.

ConclusionsIn the setting of whole blood, human CD14hiCD16+monocytes are the most efficient subset at ingestingantibody-opsonised IE. This is not observed in assaysusing isolated PBMC where classical and intermediatemonocytes show similar phagocytosis of antibody-opsonised IE. In whole blood, phagocytosis of IgG-opsonised IE requires Fcγ receptor IIIa but not otherFcγ receptors. However Fcγ receptor IIIa is not able tomediate phagocytosis of IgG-opsonised IE on its ownbut requires complement opsonisation of IE. Assaysmeasuring IE phagocytosis using PBMC or purifiedmonocyte preparations therefore do not detect criticalelements of antibody-mediated phagocytosis. We con-clude that assays which measure the ability of vaccinesto elicit a protective antibody response to P. falciparumshould consider their ability to promote phagocytosisand fix complement.

Additional files

Additional file 1: Phagocytosis of E.coli by monocyte subsetsassayed in whole blood. Whole blood from a single donor (100 μL) wasincubated with 2 x 108 pHrodo® Red-labelled Escherichia coli BioParticleconjugates (Life Technologies) for 10 minutes at either 4 °C or 37 °C asindicated, then the cells were stained with CD14-APC and CD16-FITC,and uptake of E.coli determined by flow cytometry. The percentage ofclassical (C; black bars), intermediate (IM; white bars) and non-classical(NC; grey bars) monocytes that had ingested E.coli are plotted.

Zhou et al. BMC Medicine (2015) 13:154 Page 13 of 14

Additional file 2: Effect of red blood cells and autologous plasmaon phagocytosis of antibody-opsonised IE by individual monocytesubsets. (A) PBMC were mixed with O- erythrocytes at the indicated cellratios then phagocytosis of CS2-IE opsonised with rabbit anti human RBCantibody was measured as described in Methods. Percent phagocytosisof IE by each subset is expressed relative to the degree of phagocytosisby the same subset in PBMC alone (zero RBC). Intermediate monocytes (IM),open circles; classical monocytes (C), closed circles; non-classical monocytes(NC), open squares. Data represent mean (sem) of four independentexperiments. (B) PBMC washed in PBS containing 2 % FBS were resuspendedin PBS containing the indicated final concentrations of autologous plasma,then phagocytosis of CS2-IE opsonised with rabbit anti human RBC antibodywas measured as described in Methods. Percent phagocytosis of each subsetdefined as in (A) is expressed relative to phagocytosis by the same subset inPBMC alone (zero added autologous plasma). Data represent mean (sem) offour independent experiments.

Additional file 3: AMNIS Image Stream flow cytometric analysisof phagocytosis of antibody-opsonised IE by monocytes.Phagocytosis of PKH26-labelled CS2-IE opsonised with rabbit antihuman RBC antibody by monocyte subsets was analysed by imagingflow cytometry as described in Methods. (A) Gating strategy used toexclude events that were not in focus (left panel) or cell doublets(middle panel) and to identify the individual monocyte subsets (right panel).(B) Representative images of a classical monocyte (C: CD14hiCD16-; top panel),an intermediate monocyte (IM: CD14hiCD16+; middle panel) and anon-classical monocyte (NC: CD14loCD16+; lower panel) from the respectivegates defined in (A) above. (C) Representative images of three monocyteswithin the intermediate monocyte gate containing ingested parasites(indicated on the bright field image by arrows) and showing co-localisationof CD16 and PKH26. Due to the differences in brightness of the fluorophoresused, it was difficult to depict CD14 on the co-localisation panels; however,all monocytes within the intermediate monocyte gate stained positive forboth CD14 and CD16 as shown in the representative cell illustrated in the IMpanels in (B). The numbers within the bright field images in Fig. 2c refer tothe event number assigned by the IDEAS software.

Additional file 4: Inhibition of iRBC phagocytosis following additionof heat inactivated serum and Ig-depleted heat-inactivated serum.A total of 0.5 mL of human serum was heat inactivated for 30 minutes at56 °C. One half of the serum was depleted of immunoglobulins byincubation with 125 μL Protein G sepharose (Pharmacia) for 20 minutesat room temperature with constant mixing, then the Protein G wasremoved by centrifugation. PBMC (0.5 x 106) were incubated with 1 x 107

IE (CS2-infected RBC opsonised with rabbit anti-human IgG) in thepresence of either 0 or 50 % treated and untreated serum. Phagocytosisby classical (C), intermediate (IM and non-classical (NC) monocyte subsetswas measured as described in Methods. The percent phagocytosis of eachsubset is expressed as a percentage of that of the same subset measuredin the absence of added serum (0). Black bars: heat inactivated serum.Open bars; Ig-depleted, heat-inactivated serum.

AbbreviationsADCI: antibody-dependent cellular inhibition; C3: complement component 3;C3b: complement component 3b; CR: complement receptor; EtBr: ethidiumbromide; FACS: fluorescence activated cell sorting; FBS: foetal bovine serum;iC3b: inactive C3b; ICAM-1: intracellular adhesion molecule-1; IE: infectederythrocytes; iRBC: infected red blood cells; ITAM: immunoreceptor tyrosine-based activation motif; PBMC: peripheral blood mononuclear cells;PBS: phosphate buffered saline; PfEMP-1: plasmodium falciparum erythrocytemembrane protein 1; PI: phagocytic index (number of ingested particles per100 monocytes); RBC: red blood cells; TLR: toll-like receptor; TNF: tumournecrosis factor.

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsJZ performed the research with help from GF and analysed the data. YYperformed research and helped analyse and interpret the image stream data.AJ designed the research and analysed the data and wrote the manuscript

with help from SR. MH and JB provided vital reagents and contributed tothe manuscript. All authors read and approved the final manuscript.

AcknowledgementsRBCs and serum for parasite culture were provided by the Red Cross BloodBank (Melbourne, Australia). Funding was provided by the National Healthand Medical Research Council of Australia (Project grant to SJR and AJ,Program grant to JB; Research Fellowship to JB). The Burnet Institute issupported by the Victorian State Government Operational InfrastructureSupport grant, and NHMRC Infrastructure for Research Institutes SupportScheme Grant.

Author details1Centre for Biomedical Research, Burnet Institute, Melbourne, Victoria 3004,Australia. 2Department of Medicine, University of Melbourne, Melbourne,Victoria 3050, Australia. 3Department of Chemical and BiomolecularEngineering, University of Melbourne, Melbourne, Victoria 3800, Australia.4Department of Infectious Diseases, Monash University, Melbourne, Victoria3800, Australia. 5Department of Immunology, Monash University, Melbourne,Victoria 3800, Australia. 6Department of Microbiology, Monash University,Melbourne, Victoria 3800, Australia.

Received: 25 February 2015 Accepted: 3 June 2015

References1. World Health Organization, World Malaria Report 2013. http://www.who.int/

malaria/publications/world_malaria_report_2013/report/en/ accessed 26/6/2015

2. Desai M, ter Kuile FO, Nosten F, McGready R, Asamoa K, Brabin B, et al.Epidemiology and burden of malaria in pregnancy. Lancet Infect Dis.2007;7:93–104.

3. Rogerson SJ, Hviid L, Duffy PE, Leke RF, Taylor DW. Malaria in pregnancy:pathogenesis and immunity. Lancet Infect Dis. 2007;7:105–17.

4. Marsh K, Kinyanjui S. Immune effector mechanisms in malaria. ParasiteImmunol. 2006;28:51–60.

5. Chan JA, Fowkes FJ, Beeson JG. Surface antigens of Plasmodium falciparum-infected erythrocytes as immune targets and malaria vaccine candidates.Cell Mol Life Sci. 2014;71:3633–57.

6. Chan JA, Howell KB, Reiling L, Ataide R, Mackintosh CL, Fowkes FJ, et al.Targets of antibodies against Plasmodium falciparum-infected erythrocytesin malaria immunity. J Clin Invest. 2012;122:3227–38.

7. Salanti A, Dahlback M, Turner L, Nielsen MA, Barfod L, Magistrado P, et al.Evidence for the involvement of VAR2CSA in pregnancy-associated malaria.J Exp Med. 2004;200:1197–203.

8. Tutterrow YL, Salanti A, Avril M, Smith JD, Pagano IS, Ako S, et al. Highavidity antibodies to full-length VAR2CSA correlate with absence ofplacental malaria. PLoS One. 2012;7:e40049.

9. Tutterrow YL, Avril M, Singh K, Long CA, Leke RJ, Sama G, et al. High levelsof antibodies to multiple domains and strains of VAR2CSA correlate withthe absence of placental malaria in Cameroonian women living in anarea of high Plasmodium falciparum transmission. Infect Immun.2012;80:1479–90.

10. Mayor A, Kumar U, Bardaji A, Gupta P, Jimenez A, Hamad A, et al. Improvedpregnancy outcomes in women exposed to malaria with high antibodylevels against Plasmodium falciparum. J Infect Dis. 2013;207:1664–74.

11. Chandrasiri UP, Randall LM, Saad AA, Bashir AM, Rogerson SJ, Adam I. Lowantibody levels to pregnancy-specific malaria antigens and heightenedcytokine responses associated with severe malaria in pregnancy. J Infect Dis.2014;209:1408–17.

12. Vernes A. Phagocytosis of P falciparum parasitised erythrocytes byperipheral monocytes. Lancet. 1980;2:1297–8.

13. Galbraith RM, Fox H, Hsi B, Galbraith GM, Bray RS, Faulk WP. The humanmaterno-foetal relationship in malaria. II. Histological, ultrastructuraland immunopathological studies of the placenta. Trans R Soc TropMed Hyg. 1980;74:61–72.

14. Yamada M, Steketee R, Abramowsky C, Kida M, Wirima J, Heymann D, et al.Plasmodium falciparum associated placental pathology: a light and electronmicroscopic and immunohistologic study. Am J Trop Med Hyg.1989;41:161–8.

Zhou et al. BMC Medicine (2015) 13:154 Page 14 of 14

15. Ordi J, Ismail MR, Ventura PJ, Kahigwa E, Hirt R, Cardesa A, et al. Massivechronic intervillositis of the placenta associated with malaria infection. Am JSurg Pathol. 1998;22:1006–11.

16. Celada A, Cruchaud A, Perrin LH. Opsonic activity of human immune serumon in vitro phagocytosis of Plasmodium falciparum infected red blood cellsby monocytes. Clin Exp Immunol. 1982;47:635–44.

17. Turrini F, Ginsburg H, Bussolino F, Pescarmona GP, Serra MV, Arese P.Phagocytosis of Plasmodium falciparum-infected human red blood cells byhuman monocytes: involvement of immune and nonimmune determinantsand dependence on parasite developmental stage. Blood. 1992;80:801–8.

18. Wahlgren M, Abrams JS, Fernandez V, Bejarano MT, Azuma M, Torii M, et al.Adhesion of Plasmodium falciparum-infected erythrocytes to human cellsand secretion of cytokines (IL-1-beta, IL-1RA, IL-6, IL-8, IL-10, TGF beta, TNFalpha, G-CSF, GM-CSF. Scand J Immunol. 1995;42:626–36.

19. Nagao T, Uemura H, Yanagi T, Oishi K, Nagatake T, Kanbara H. Loss of tumornecrosis factor production by human monocytes in falciparum malaria aftertheir maturation in vitro. Am J Trop Med Hyg. 1996;55:562–6.

20. Ziegler-Heitbrock L, Ancuta P, Crowe S, Dalod M, Grau V, Hart DN, et al.Nomenclature of monocytes and dendritic cells in blood. Blood.2010;116:e74–80.

21. Tacke F, Alvarez D, Kaplan TJ, Jakubzick C, Spanbroek R, Llodra J, et al.Monocyte subsets differentially employ CCR2, CCR5, and CX3CR1 toaccumulate within atherosclerotic plaques. J Clin Invest. 2007;117:185–94.

22. Cros J, Cagnard N, Woollard K, Patey N, Zhang SY, Senechal B, et al. HumanCD14dim monocytes patrol and sense nucleic acids and viruses via TLR7and TLR8 receptors. Immunity. 2010;33:375–86.

23. Jaworowski A, Fernandes LA, Yosaatmadja F, Feng G, Mwapasa V, MolyneuxME, et al. Relationship between human immunodeficiency virus type 1coinfection, anemia, and levels and function of antibodies to variant surfaceantigens in pregnancy-associated malaria. Clin Vaccine Immunol.2009;16:312–9.

24. Feng G, Aitken E, Yosaatmadja F, Kalilani L, Meshnick SR, Jaworowski A, et al.Antibodies to variant surface antigens of Plasmodium falciparum-infectederythrocytes are associated with protection from treatment failure and thedevelopment of anemia in pregnancy. J Infect Dis. 2009;200:299–306.

25. Ataide R, Mwapasa V, Molyneux ME, Meshnick SR, Rogerson SJ. Antibodiesthat induce phagocytosis of malaria infected erythrocytes: effect of HIVinfection and correlation with clinical outcomes. PLoS One. 2011;6:e22491.

26. Osier FH, Feng G, Boyle MJ, Langer C, Zhou J, Richards JS, et al. Opsonicphagocytosis of Plasmodium falciparum merozoites: mechanism in humanimmunity and a correlate of protection against malaria. BMC Med.2014;12:108.

27. Tippett E, Fernandes LA, Rogerson SJ, Jaworowski A. A novel flowcytometric phagocytosis assay of malaria-infected erythrocytes. J ImmunolMethods. 2007;325:42–50.

28. Umbers AJ, Boeuf P, Clapham C, Stanisic DI, Baiwog F, Mueller I, et al.Placental malaria-associated inflammation disturbs the insulin-like growthfactor axis of fetal growth regulation. J Infect Dis. 2011;203:561–9.

29. Cooke BM, Rogerson SJ, Brown GV, Coppel RL. Adhesion of malaria-infectedred blood cells to chondroitin sulfate A under flow conditions. Blood.1996;88:4040–4.

30. Biggs BA, Anders RF, Dillon HE, Davern KM, Martin M, Petersen C,et al. Adherence of infected erythrocytes to venular endotheliumselects for antigenic variants of Plasmodium falciparum. J Immunol.1992;149:2047–54.

31. Hommel M, Elliott SR, Soma V, Kelly G, Fowkes FJ, Chesson JM, et al.Evaluation of the antigenic diversity of placenta-binding Plasmodiumfalciparum variants and the antibody repertoire among pregnantwomen. Infect Immun. 2010;78:1963–78.

32. Ramsland PA, Farrugia W, Bradford TM, Sardjono CT, Esparon S, Trist HM,et al. Structural basis for Fc gammaRIIa recognition of human IgG and formationof inflammatory signaling complexes. J Immunol. 2011;187:3208–17.

33. Duffy MF, Byrne TJ, Carret C, Ivens A, Brown GV. Ectopic recombination of amalaria var gene during mitosis associated with an altered var switchrate. J Mol Biol. 2009;389:453–69.

34. Van Geertruyden JP, Van Eijk E, Yosaatmadja F, Kasongo W, Mulenga M,D”Alessandro U, et al. The relationship of Plasmodium falciparum humeralimmunity with HIV-1 immunosuppression and treatment efficacy in Zambia.Malar J. 2009;8:258.

35. Struik SS, Omer FM, Artavanis-Tsakonas K, Riley EM. Uninfected erythrocytesinhibit Plasmodium falciparum-induced cellular immune responses inwhole-blood assays. Blood. 2004;103:3084–92.

36. Facer CA, Bray RS, Brown J. Direct Coombs antiglobulin reactions inGambian children with Plasmodium falciparum malaria. I. Incidence andclass specificity. Clin Exp Immunol. 1979;35:119–27.

37. Preynat-Seauve O, Villiers CL, Jourdan G, Richard MJ, Plumas J, Favier A,et al. An interaction between CD16 and CR3 enhances iC3b binding to CR3but is lost during differentiation of monocytes into dendritic cells. Eur JImmunol. 2004;34:147–55.

38. Li X, Baskin JG, Mangan EK, Su K, Gibson AW, Ji C, et al. The uniquecytoplasmic domain of human FcgammaRIIIA regulates receptor-mediatedfunction. J Immunol. 2012;189:4284–94.

39. Khusmith S, Druilhe P. Antibody-dependent ingestion of P. falciparummerozoites by human blood monocytes. Parasite Immunol. 1983;5:357–68.

40. Groux H, Gysin J. Opsonization as an effector mechanism in humanprotection against asexual blood stages of Plasmodium falciparum:functional role of IgG subclasses. Res Immunol. 1990;141:529–42.

41. Hogarth PM, Pietersz GA. Fc receptor-targeted therapies for the treatmentof inflammation, cancer and beyond. Nat Rev Drug Discov. 2012;11:311–31.

42. Cooke GS, Aucan C, Walley AJ, Segal S, Greenwood BM, Kwiatkowski DP,et al. Association of Fcgamma receptor IIa (CD32) polymorphism withsevere malaria in West Africa. Am J Trop Med Hyg. 2003;69:565–8.

43. Ouma C, Keller CC, Opondo DA, Were T, Otieno RO, Otieno MF, et al.Association of FCgamma receptor IIA (CD32) polymorphism with malarialanemia and high-density parasitemia in infants and young children. Am JTrop Med Hyg. 2006;74:573–7.

44. Nasr A, Iriemenam NC, Troye-Blomberg M, Giha HA, Balogun HA, Osman OF,et al. Fc gamma receptor IIa (CD32) polymorphism and antibody responsesto asexual blood-stage antigens of Plasmodium falciparum malaria inSudanese patients. Scand J Immunol. 2007;66:87–96.

45. Ouma C, Davenport GC, Garcia S, Kempaiah P, Chaudhary A, Were T, et al.Functional haplotypes of Fc gamma (Fcgamma) receptor (FcgammaRIIAand FcgammaRIIIB) predict risk to repeated episodes of severe malarialanemia and mortality in Kenyan children. Hum Genet. 2012;131:289–99.

46. Wu J, Edberg JC, Redecha PB, Bansal V, Guyre PM, Coleman K, et al. A novelpolymorphism of FcgammaRIIIa (CD16) alters receptor function andpredisposes to autoimmune disease. J Clin Invest. 1997;100:1059–70.

47. Lassauniere R, Shalekoff S, Tiemessen CT. A novel FCGR3A intragenichaplotype is associated with increased FcgammaRIIIa/CD16a cell surfacedensity and population differences. Hum Immunol. 2013;74:627–34.

48. Baranji K, Baranyi L, Yoshida T, Okada N, Okada H. A low molecular weightphagocytosis-inhibitory factor obtained from human erythrocyte membranesspecifically down-regulates Mac-1 activity on tetradecanoyl phorbol acetate-stimulated monocytic cell lines in a Ca(2+)-dependent manner. J Immunol.1994;153:1724–35.

49. Chimma P, Roussilhon C, Sratongno P, Ruangveerayuth R, PattanapanyasatK, Perignon JL, et al. A distinct peripheral blood monocyte phenotype isassociated with parasite inhibitory activity in acute uncomplicatedPlasmodium falciparum malaria. PLoS Pathog. 2009;5, e1000631.

50. Antonelli LR, Leoratti FM, Costa PA, Rocha BC, Diniz SQ, Tada MS,et al. The CD14+CD16+ inflammatory monocyte subset displaysincreased mitochondrial activity and effector function during acutePlasmodium vivax malaria. PLoS Pathog. 2014;10. e1004393.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit